| { | |
| "run": { | |
| "task": "video_text_pretrain", | |
| "lr_sched": "linear_warmup_cosine_lr", | |
| "init_lr": 3e-05, | |
| "min_lr": 1e-05, | |
| "warmup_lr": 1e-06, | |
| "weight_decay": 0.05, | |
| "max_epoch": 3, | |
| "iters_per_epoch": 51377, | |
| "batch_size_train": 1, | |
| "batch_size_eval": 4, | |
| "num_workers": 4, | |
| "warmup_steps": 25688, | |
| "seed": 42, | |
| "output_dir": "/data/mjjung/Video-LLaMA/videollama_activitynet_vtune", | |
| "amp": true, | |
| "resume_ckpt_path": null, | |
| "evaluate": false, | |
| "train_splits": [ | |
| "train" | |
| ], | |
| "device": "cuda", | |
| "world_size": 4, | |
| "dist_url": "env://", | |
| "distributed": true, | |
| "rank": 0, | |
| "gpu": 0, | |
| "dist_backend": "nccl" | |
| }, | |
| "model": { | |
| "arch": "video_llama", | |
| "image_size": 224, | |
| "drop_path_rate": 0, | |
| "use_grad_checkpoint": false, | |
| "vit_precision": "fp16", | |
| "freeze_vit": true, | |
| "freeze_qformer": true, | |
| "num_query_token": 32, | |
| "llama_model": "/data/mjjung/vicuna-7b-v1.5", | |
| "prompt": "", | |
| "model_type": "pretrain_vicuna", | |
| "imagebind_ckpt_path": "/data/mjjung/Video-LLaMA/Video-LLaMA-2-7B-Finetuned/", | |
| "ckpt": "/data/mjjung/Video-LLaMA/Video-LLaMA-2-7B-Finetuned/VL_LLaMA_2_7B_Finetuned.pth", | |
| "equip_audio_branch": false, | |
| "frozen_llama_proj": false, | |
| "frozen_video_Qformer": false, | |
| "frozen_audio_Qformer": true, | |
| "fusion_head_layers": 2, | |
| "max_frame_pos": 32, | |
| "fusion_header_type": "seqTransf", | |
| "max_txt_len": 320, | |
| "end_sym": "###", | |
| "prompt_path": "", | |
| "prompt_template": "###Human: {} ###Assistant: " | |
| }, | |
| "preprocess": { | |
| "vis_processor": { | |
| "train": { | |
| "name": "alpro_video_train", | |
| "image_size": 224, | |
| "n_frms": 8 | |
| }, | |
| "eval": { | |
| "name": "alpro_video_eval", | |
| "image_size": 224, | |
| "n_frms": 8 | |
| } | |
| }, | |
| "text_processor": { | |
| "train": { | |
| "name": "blip_caption" | |
| }, | |
| "eval": { | |
| "name": "blip_caption" | |
| } | |
| } | |
| }, | |
| "datasets": { | |
| "webvid_instruct": { | |
| "data_type": "video", | |
| "build_info": { | |
| "anno_dir": "data/activitynet_filtered_qa_grounding.json", | |
| "videos_dir": "/data/video_datasets/" | |
| }, | |
| "vis_processor": { | |
| "train": { | |
| "name": "alpro_video_train", | |
| "n_frms": 8, | |
| "image_size": 224 | |
| } | |
| }, | |
| "text_processor": { | |
| "train": { | |
| "name": "blip_caption" | |
| } | |
| }, | |
| "num_video_query_token": 32, | |
| "tokenizer_name": "/data/mjjung/vicuna-7b-v1.5", | |
| "model_type": "vicuna" | |
| } | |
| } | |
| } | |
| {"train_lr": "0.000", "train_loss": "0.230"} | |
| {"train_lr": "0.000", "train_loss": "0.204"} | |